Siemens Deploys AI to Speed Chip Design
Siemens announced it is using agentic AI in its Questa One platform to accelerate the design and verification of integrated circuits. The company claims the AI-driven workflows will speed up the register-transfer level (RTL) sign-off process, a critical step in semiconductor development, while integrating with customers' existing investments.
The new "Questa One Agentic Toolkit" from Siemens features five distinct AI agents. These include an RTL Code Agent for generating synthesizable code, a Lint Agent for checking design errors, and agents dedicated to clock domain crossing (CDC) verification, automated verification planning, and debugging. This AI system targets the register-transfer level (RTL) sign-off, a critical phase in chip development. RTL sign-off focuses on the structural correctness of a design, ensuring the code is ready for physical implementation and free of issues that could derail the synthesis or timing. It serves as a final quality check on the digital blueprint before committing to manufacturing. To power these AI agents, Siemens is collaborating with NVIDIA, leveraging NVIDIA NIM and Nemotron models. This partnership highlights a trend of EDA (electronic design automation) companies using foundational AI models from major tech players to build specialized, industrial-grade tools. Early feedback from partners indicates rapid adoption. Akshay Aggarwal, a senior engineering director at MediaTek, reported that their engineers achieved proficiency with the toolkit within hours and saw "immediate and significant" productivity gains. The move is part of a larger industry race to integrate agentic AI into chip design. Competitor Cadence recently launched its ChipStack AI Super Agent, also claiming productivity improvements of up to 10x by automating tasks like testbench generation and failure analysis. This push toward AI automation is a direct response to the escalating complexity of semiconductor design and a shortage of specialized engineers. As chip designs for AI and high-performance computing become more intricate, companies are turning to AI agents to manage the workload and accelerate time-to-market.
